The mental model theory postulates that reasoners build models of the situations described in premises, and that these models normally represent only what is true. The theory has an unexpected consequence. It predicts the existence ofillusions in inferences. Certain inferences should have compelling but erroneous conclusions. Two experiments corroborated the occurrence of such illusions in inferences about what is possible from disjunctions of quantified assertions, such as, “at least some of the plastic beads are not red.” Experiment 1 showed that (...) participants erroneously inferred that impossible situations were possible, and that possible situations were impossible, but that they performed well with control problems based on the same premises. Experiment 2 corroborated these findings in inferences from assertions based on dyadic relations, such as, “all the boys played with the girls.”. (shrink)

It has been reported as a robust effect that people are likely to select a matching case in the Wason selection task. For example, they usually select the 5 case, in the Wason selection task with the conditional "if an E, then a not-5". This was explained by the matching bias account that people are likely to regard a matching case as relevant to the truth of the conditional (Evans, 1998). However, because a positive concept usually constructs a smaller set (...) than its negative one does (a rarity assumption), it is more effective to get information on the truth of the conditional in a positive set than in a negative set. Thus the optimal data selection account can also explain the effect. The set size of Q and matching by introducing negation were manipulated independently in four experiments. From the results it was inferred that the so-called matching bias was an amalgam of two different cognitive components-relevance judgement by matching and optimal data selection. (shrink)

Yama (2001) argued that the matching bias effect was an amalgam of a genuine matching effect and a set size effect. However, Evans (2002) pointed out some problems with Yama's argument. One was on the matching index used in Yama (2001). The others concerned a gap between Yama's (2001) interpretation of matching bias and Evans' (1998) recent discussions that had placed emphasis on the problem of implicit negation. I reply to these claims taking Oaksford (2002) into consideration. Further, I propose (...) a framework connecting context, goal, utility, and relevance, and including a hierarchical structure of goals. I argue that theories of reasoning can be developed and extended with the framework. (shrink)

In this letter, we argue that an economic perspective on the mind has played—and should continue to play—a central role in the development of cognitive science. Viewing cognition as the productive application of mental resources puts cognitive science and economics on a common conceptual footing, paving the way for closer collaboration between the two disciplines. This will enable cognitive scientists to more readily repurpose economic concepts and analytical tools for the study of mental phenomena, while at the same time, enriching (...) our understanding of the modern economy, which is increasingly driven by mental, rather than physical, production. (shrink)

In his commentary, Oaksford makes two main claims: (1) that the externalisation method used by Green, Over, and Pyne (1997) enforces the correlation observed between probability estimates and selection, and (2) that these estimates support the prediction of a downward revision of P(p) when P(p) > P(q). In this reply, we rebut claim 1 by describing the instructions more comprehensively, and claim 2 by reiterating the importance of making certain theoretical distinctions which Oaksford does not make. Our interest is the (...) psychological process of reaching decision: externalisation methods provide a means of exploring this process and of assessing the value of Bayesian approaches. (shrink)

We do not all make choices, reason, interpret our experience, or respond to our environment in the same way. A recent surge of scientific interest has thrust these individual differences into the spotlight: researchers in cognitive psychology and neuroscience are now devoting increasing attention to cognitive variation. The philosophical dimensions of this research, however, have yet to be systematically explored. Here I make an initial foray by considering how cognitive variation is characterized. I present a central dilemma facing descriptions of (...) individual differences, discuss several distinctions used to categorize variation, and show that these distinctions require further elaboration. Finally, by canvassing several philosophical topics for which the characterization of cognitive variation may have significant implications, I argue that philosophers should take note of how and why our minds differ. (shrink)

On appelle « biais de confirmation » la tendance supposée des humains à sélectionner les informations qui vont dans le sens de ce qu’ils croient (ou veulent croire) et à interpréter celles dont ils disposent en faveur de leurs hypothèses favorites. Cet article vise à porter un regard critique sur certains usages de cette notion, et plus généralement sur le recours aux « biais cognitifs » pour expliquer tout un ensemble de phénomènes sociaux interprétés comme les marques d’une forme d’irrationalité. (...) Remettant en question le caractère probant des résultats expérimentaux destinés à mettre en évidence une déviation qui nous conduirait systématiquement à des représentations inexactes, je suggère que l’approche bayésienne en psychologie du raisonnement permet de rendre compte de certains de ces résultats en faisant l’hypothèse que nos inférences se conforment à une telle norme. (shrink)

ABSTRACTProbability judgements entail a conjunction fallacy if a conjunction is estimated to be more probable than one of its conjuncts. In the context of predication of alternative logical hypothesis, Bayesian logic provides a formalisation of pattern probabilities that renders a class of pattern-based CFs rational. BL predicts a complete system of other logical inclusion fallacies. A first test of this prediction is investigated here, using transparent tasks with clear set inclusions, varying in observed frequencies only. Experiment 1 uses data where (...) BL makes dominant predictions; Experiment 2's predictions were less clear, and we additionally investigated judgements about the second-most probable hypotheses. The results corroborated a pattern-probability account and cannot be easily explained by other theories of CFs. IFs were not limited to conjunctions, but rather occurred systematically for several logical connectives. Thus, pattern-based prob... (shrink)

In evolutionary biology, a trait is said to be optimal if it maximizes the fitness of the organism, that is, if the trait allows the organism to survive and reproduce better than any other competing trait would. In engineering, a design is said to be optimal if it complies with its functional requirements as well as possible. Cognitive science is both a biological and engineering discipline and hence it uses both notions of optimality. Unfortunately, the lack of a clear methodological (...) stance on this tissue has made it common for researchers to conflate these two kinds of optimality. In this paper, I argue that a strict distinction must kept in order to avoid inaccurate assumptions. (shrink)

Some studies have reported that, under some circumstances, participants sometimes reject the truth of conditional premises and give incorrect uncertain conclusions to MP and MT, despite the standard instructions to assume the truth of the premises. Instructions that emphasise the logical nature of the task, on the other hand, increase the number of valid conclusions to these two inferences. In this paper, we examine two possible explanations for the influence of instructions on the production of valid conclusions: (1) instructions trigger (...) a logical mode of thinking that is appropriate for the task to be solved, and (2) instructions affect the way that retrieved information is used when constructing a mental representation of the premises. In order to compare these two hypothesis, we used conditional causal relations having either many or few disabling conditions (i.e. cases of P and not-Q) and presented them in four different instructional conditions. When presented with the standard instructions to suppose the premises to be true, MP and MT inferences show a relatively low percentage of valid conclusions. However, when the logical nature of the task and the necessity to assume the truth of the premises are emphasised, the number of valid conclusions greatly increases. Also, if disabling conditions are made explicit or generated before the presentation of these logical instructions, the performance on MP and MT is lower than when only the logical instructions are presented. Finally, the instructions had no effect on AC and DA. From these results, we concluded that the instructions had an effect on the use of retrieved information rather than on inducing different modes of thinking. (shrink)

In recent years, a multitude of datasets of human–human conversations has been released for the main purpose of training conversational agents based on data‐hungry artificial neural networks. In this paper, we argue that datasets of this sort represent a useful and underexplored source to validate, complement, and enhance cognitive studies on human behavior and language use. We present a method that leverages the recent development of powerful computational models to obtain the fine‐grained annotation required to apply metrics and techniques from (...) Cognitive Science to large datasets. Previous work in Cognitive Science has investigated the question‐asking strategies of human participants by employing different variants of the so‐called 20‐question‐game setting and proposing several evaluation methods. In our work, we focus on GuessWhat, a task proposed within the Computer Vision and Natural Language Processing communities that is similar in structure to the 20‐question‐game setting. Crucially, the GuessWhat dataset contains tens of thousands of dialogues based on real‐world images, making it a suitable setting to investigate the question‐asking strategies of human players on a large scale and in a natural setting. Our results demonstrate the effectiveness of computational tools to automatically code how the hypothesis space changes throughout the dialogue in complex visual scenes. On the one hand, we confirm findings from previous work on smaller and more controlled settings. On the other hand, our analyses allow us to highlight the presence of “uninformative” questions (in terms of Expected Information Gain) at specific rounds of the dialogue. We hypothesize that these questions fulfill pragmatic constraints that are exploited by human players to solve visual tasks in complex scenes successfully. Our work illustrates a method that brings together efforts and findings from different disciplines to gain a better understanding of human question‐asking strategies on large‐scale datasets, while at the same time posing new questions about the development of conversational systems. (shrink)

Current psychology of human reasoning is divided into several different approaches. For instance, there is a major dispute over the question whether human beings are able to apply norms of the formal models of rationality such as rules of logic, or probability and decision theory, correctly. While researchers following the “heuristics and biases” approach argue that we deviate systematically from these norms, and so are perhaps deeply irrational, defenders of the “bounded rationality” approach think not only that the evidence for (...) this conclusion is problematic but also that we should not, at least not very often, use formal norms in reasoning. I argue that while the evidence for heuristics and biases is indeed questionable, the bounded rationality approach has its limits too. Most especially, we should not infer that formal norms play no role in a comprehensive theory of rationality. Instead, formal and bounded rules of reasoning might even be connected in a more comprehensive theory of rationality. (shrink)

We review the various explanations that have been offered toaccount for subjects'' behaviour in Wason ''s famous selection task. Weargue that one element that is lacking is a good understanding ofsubjects'' semantics for the key expressions involved, and anunderstanding of how this semantics is affected by the demands the taskputs upon the subject''s cognitive system. We make novel proposals inthese terms for explaining the major content effects of deonticmaterials. Throughout we illustrate with excerpts from tutorialdialogues which motivate the kinds of (...) analysis proposed. Our long termgoal is an integration of the various insights about conditionalreasoning on offer from different cognitive science methodologies. Thepurpose of this paper is to try to draw the attention of logicians andsemanticists to this area, since we believe that empirical investigationof the cognitive processes involved could benefit from semanticanalyses. (shrink)

In this response, we clarify several misunderstandings of the understanding/acceptance principle and defend our specific operationalization of that principle. We reiterate the importance of addressing the problem of rational task construal and we elaborate the notion of computational limitations contained in our target article. Our concept of thinking dispositions as variable intentional-level styles of epistemic and behavioral regulation is explained, as is its relation to the rationality debate. Many of the suggestions of the commentators for elaborating two-process models are easily (...) integrated into our generic dual-process account. We further explicate how we view the relation between System 1 and System 2 and evolutionary and normative rationality. We clarify our attempt to fuse the contributions of the cognitive ecologists with the insights of the original heuristics and biases researchers. (shrink)

Suppose that two competing norms, N1 and N2, can be identified such that a given person’s response can be interpreted as correct according to N1 but incorrect according to N2. Which of these two norms, if any, should one use to interpret such a response? In this paper we seek to address this fundamental problem by studying individual variation in the interpretation of conditionals by establishing individual profiles of the participants based on their case judgments and reflective attitudes. To investigate (...) the participants’ reflective attitudes we introduce a new experimental paradigm called the Scorekeeping Task. As a case study, we identify the participants who follow the Suppositional Theory of conditionals (N1) versus Inferentialism (N2) and investigate to what extent internally consistent competence models can be reconstructed for the participants on this basis. After extensive empirical investigations, an apparent reasoning error with and-to-if inferences was found in one of these two groups. The implications of this case study for debates on the proper role of normative considerations in psychology are discussed. (shrink)

Quite unexpectedly, cognitive psychologists find their field intimately connected to a whole new intellectual landscape that had previously seemed remote, unfamiliar, and all but irrelevant. Yet the proliferating connections tying together the cognitive and evolutionary communities promise to transform both fields, with each supplying necessary principles, methods, and a species of rigor that the other lacks. (Cosmides and Tooby, 1994, p. 85).

We present a meta-analytic review on the processing of negations in conditional reasoning about affirmation problems (Modus Ponens: “MP”, Affirmation of the Consequent “AC”) and denial problems (Denial of the Antecedent “DA”, and Modus Tollens “MT”). Findings correct previous generalisations about the phenomena. First, the effects of negation in the part of the conditional about which an inference is made, are not constrained to denial problems. These inferential-negation effects are also observed on AC. Second, there generally are reliable effects of (...) a negation in the clause referred to by the categorical premise, and these referred-negation effects are constrained to the logically fallacious AC and DA inferences. All findings are presented and discussed in relation to contemporary mental model (MM) and mental logic (ML) theories. It is argued that a double-negation elimination hypothesis provides a sufficient explanation of inferential-negation effects within both MM theory and ML theory, if the latter is extended by a validating search for counter examples. Both MM and ML theories adhere to a processing scheme that allows them to incorporate an account of referred-negation effects based on the thesis that counter-example frequency is modulated by the scope of a contrast class delineated by a false affirmative. We conclude that MM and ML theories provide adequate processing schemes to accommodate for the explanatory hypotheses, at least in principle. In practice, both approaches remain equivocal as regards the connectivity and interactivity with long-term memory knowledge invoked in generating, manipulating, and testing the mental representations of negative state of affairs. (shrink)

One of the most discussed issues in psychology—presently and in the past—is how to define and measure the extent to which human cognition is rational. The rationality of human cognition is often evaluated in terms of normative standards based on a priori intuitions. Yet this approach has been challenged by two recent developments in psychology that we review in this article: ecological rationality and descriptivism. Going beyond these contributions, we consider it a good moment for psychologists and philosophers to join (...) forces and work toward a new foundation for the definition of rational cognition. We take a first step in this direction by proposing that the rationality of both cognitive and normative systems can be measured in terms of their cognitive success. Cognitive success can be defined and gauged in terms of two factors: ecological validity (the system's validity in conditions in which it is applicable) and the system's applicability (the scope of conditions under which it can be applied). As we show, prominent systems of reasoning—deductive reasoning, Bayesian reasoning, uncertain conditionals, and prediction and choice—perform rather differently on these two factors. Furthermore, we demonstrate that conceptualizing rationality according to its cognitive success offers a new perspective on the time-honored relationship between the descriptive (“is”) and the normative (“ought”) in psychology and philosophy. (shrink)

Samuels and Stich explore the debate over the extent to which ordinary human reasoning and decision making is rational. One prominent cluster of views, often associated with the heuristics and biases tradition in psychology, maintains that human reasoning is, in important respects, normatively problematic or irrational. Samuels and Stich start by sketching some key experimental findings from this tradition and describe a range of pessimistic claims about the rationality of ordinary people that these and related findings are sometimes taken to (...) support. Such pessimistic interpretations of the experimental findings have not gone unchallenged however: Samuels and Stich outline some of the research on reasoning that has been done by evolutionary psychologists and sketch a cluster of more optimistic theses about ordinary reasoning that such psychologists defend. Although Samuels and Stich think that the most dire pronouncements made by writers in the heuristics and biases tradition are unwarranted, they also maintain that the situation is rather more pessimistic than sometimes suggested by evolutionary psychologists. They conclude by defending this “middle way” and sketch a family of “dual processing” theories of reasoning which, they argue, offer some support for the moderate interpretation they advocate. (shrink)

One way to learn about the world is by asking questions. We investigate how younger children (7- to 8-year-olds), older children (9- to 11-year-olds), and young adults (17- to 18-year-olds) ask questions to identify the cause of an event. We find a developmental shift in children’s reliance on hypothesis-scanning questions (which test hypotheses directly) versus constraint-seeking questions (which reduce the space of hypotheses), but also that all age groups ask more constraint-seeking questions when hypothesis-scanning questions are least likely to pay (...) off: When the solution is one among equally likely alternatives (Study 1) or when the problem is difficult (Studies 1 and 2). These findings are the first to demonstrate that even young children dynamically adapt their strategies for inquiry to increase the efficiency of information search. (shrink)

The main hypothesis of studies presented in this article is that episodic implicit evaluations toward task-relevant objects determine thinking and decisions by actively placing them within or outside the scope of attention. In these studies we also aimed to test the impact of regulatory focus on implicit evaluations and goal pursuit. We applied the Promotion-Prevention Self-control Scale as a measure of mind-set during thinking in the Wason Selection Task in Study 1 and Island Decision Game in Study 2. Directly after (...) learning of the tasks, participants evaluated objects that constituted the task’s content. The findings are in line with the hypothesis stating that goals influence the way in which objects are automatically evaluated. The effects of promotion mind-set were more pronounced in both studies. Promotion-focused individuals positively assessed objects that serve as a confirmation. The implicit evaluations by prevention-oriented individuals disclosed their falsifying approach to the WST. The positive implicit evaluation of correct objects suggests their sensitivity to information useful for falsification and is consistent with their tendency to cautiously self-control thinking. (shrink)

This paper focuses on the cognitive/computational and evolutionary levels. It describes three proposals to make cognition computationally tractable, namely: Resource Rationality, the Adaptive Toolbox and Massive Modularity. While each of these proposals appeals to evolutionary considerations to dissolve the intractability of cognition, Rich, Blokpoel, de Haan, and van Rooij argue that, in each case, the intractability challenge is not resolved, but just relocated to the level of evolution.

This paper focuses on the cognitive/computational and evolutionary levels. It describes three proposals to make cognition computationally tractable, namely: Resource Rationality, the Adaptive Toolbox and Massive Modularity. While each of these proposals appeals to evolutionary considerations to dissolve the intractability of cognition, Rich, Blokpoel, de Haan, and van Rooij argue that, in each case, the intractability challenge is not resolved, but just relocated to the level of evolution.

Human and automated tutors attempt to choose pedagogical activities that will maximize student learning, informed by their estimates of the student's current knowledge. There has been substantial research on tracking and modeling student learning, but significantly less attention on how to plan teaching actions and how the assumed student model impacts the resulting plans. We frame the problem of optimally selecting teaching actions using a decision-theoretic approach and show how to formulate teaching as a partially observable Markov decision process planning (...) problem. This framework makes it possible to explore how different assumptions about student learning and behavior should affect the selection of teaching actions. We consider how to apply this framework to concept learning problems, and we present approximate methods for finding optimal teaching actions, given the large state and action spaces that arise in teaching. Through simulations and behavioral experiments, we explore the consequences of choosing teacher actions under different assumed student models. In two concept-learning tasks, we show that this technique can accelerate learning relative to baseline performance. (shrink)

A historical theory of rational norms claims that, if we are supposed to think rationally, this is because it is biologically normal for us to do so. The historical theorist is committed to the view that we are supposed to think rationally only if, in the past, adult humans sometimes thought rationally. I consider whether there is any plausible model of rational norms that can be adopted by the historical theorist that is compatible with the claim that adult human beings (...) are subject to rational norms, given certain plausible empirical assumptions about our history and capabilities. I suggest that there is one such model: this model centres on the idea that a procedure is rational if it has been endorsed (or at least not rejected) by mechanisms that have the function to ensure that the subject learns to reason in a way that approaches a certain kind of optimality. (shrink)

The distinction between rules and similarity is central to our understanding of much of cognitive psychology. Two aspects of existing research have motivated the present work. First, in different cognitive psychology areas we typically see different conceptions of rules and similarity; for example, rules in language appear to be of a different kind compared to rules in categorization. Second, rules processes are typically modeled as separate from similarity ones; for example, in a learning experiment, rules and similarity influences would be (...) described on the basis of separate models. In the present article, I assume that the rules versus similarity distinction can be understood in the same way in learning, reasoning, categorization, and language, and that a unified model for rules and similarity is appropriate. A rules process is considered to be a similarity one where only a single or a small subset of an object's properties are involved. Hence, rules and overall similarity operations are extremes in a single continuum of similarity operations. It is argued that this viewpoint allows adequate coverage of theory and empirical findings in learning, reasoning, categorization, and language, and also a reassessment of the objectives in research on rules versus similarity. Key Words: categorization; cognitive explanation; language; learning; reasoning; rules; similarity. (shrink)

Topics in Cognitive Science, Volume 14, Issue 3, Page 451-466, July 2022.

This paper reviews the psychological investigation of reasoning with conditionals, putting an emphasis on recent work. In the first part, a few methodological remarks are presented. In the second part, the main theories of deductive reasoning (mental rules, mental models, and the probabilistic approach) are considered in turn; their content is summarised and the semantics they assume for if and the way they explain formal conditional reasoning are discussed, in particular in the light of experimental work on the probability of (...) conditionals. The last part presents the recent shift of interest towards the study of conditional reasoning in context, that is, with large knowledge bases and uncertain premises. (shrink)

We discuss O&C's probabilistic approach from a probability logical point of view. Specifically, we comment on subjective probability, the indispensability of logic, the Ramsey test, the consequence relation, human nonmonotonic reasoning, intervals, generalized quantifiers, and rational analysis.

We present a unified approach for investigating rational reasoning about basic argument forms involving indicative conditionals, counterfactuals, and basic quantified statements within coherence-based probability logic. After introducing the rationality framework, we present an interactive view on the relation between normative and empirical work. Then, we report a new experiment which shows that people interpret indicative conditionals and counterfactuals by coherent conditional probability assertions and negate conditionals by negating their consequents. The data support the conditional probability interpretation of conditionals and the (...) narrow-scope reading of the negation of conditionals. Finally, we argue that coherent conditional probabilities are important for probabilistic analyses of conditionals, nonmonotonic reasoning, quantified statements, and paradoxes. (shrink)

Three experiments investigated the contrasting predictions of the evolutionary and decision-theoretic approaches to deontic reasoning. Two experiments embedded a hazard management (HM) rule in a social contract scenario that should lead to competition between innate modules. A 3rd experiment used a pure HM task. Threatening material was also introduced into the antecedent, p, of a deontic rule, if p then must q. According to the evolutionary approach, more HM responses (Cosmides & Tooby, 2000) are predicted when p is threatening, whereas (...) decision theory predicts fewer. All 3 experiments were consistent with decision theory. Other theories are discussed, and it is concluded that they cannot account for the behavior observed in these experiments. (shrink)

Oaksford and Chater (1994) proposed to analyse the Wason selection task as an inductive instead of a deductive task. Applying Bayesian statistics, they concluded that the cards that participants tend to select are those with the highest expected information gain. Therefore, their choices seem rational from the perspective of optimal data selection. We tested a central prediction from the theory in three experiments: card selection frequencies should be sensitive to the subjective probability of occurrence for individual cards. In Experiment 1, (...) expected frequencies of the p- and the q-card were manipulated independently by concepts referring to large vs. small sets. Although the manipulation had an effect on card selection frequencies, there was only a weak correlation between the predicted and the observed patterns. In the second experiment, relative frequencies of individual cards were manipulated more directly by explicit frequency information. In addition, participants estimated probabilities for the four logical cases and of the conditional statement itself. The experimental manipulations strongly affected the probability estimates, but were completely unrelated to card selections. This result was replicated in a third experiment. We conclude that our data provide little support for optimal data selection theory. (shrink)

Human cognition requires coping with a complex and uncertain world. This suggests that dealing with uncertainty may be the central challenge for human reasoning. In Bayesian Rationality we argue that probability theory, the calculus of uncertainty, is the right framework in which to understand everyday reasoning. We also argue that probability theory explains behavior, even on experimental tasks that have been designed to probe people's logical reasoning abilities. Most commentators agree on the centrality of uncertainty; some suggest that there is (...) a residual role for logic in understanding reasoning; and others put forward alternative formalisms for uncertain reasoning, or raise specific technical, methodological, or empirical challenges. In responding to these points, we aim to clarify the scope and limits of probability and logic in cognitive science; explore the meaning of the explanation of cognition; and re-evaluate the empirical case for Bayesian rationality. (shrink)